Bone Morphogenetic Proteins (BMPs) act in developmental pattern formation as a paradigm of extracellular information that is spatially distributed in a gradient as a morphogen, specifying distinct cell types via morphogen levels. In vertebrate dorsal-ventral (DV) axial pattern formation the function, molecular partners, and role of each BMP component are fairly well understood, but the mechanism by which the combined function of the components form a robust BMP gradient is complex and still poorly understood. The field is at an impasse to go beyond the current paradigms and solve the mechanism by which the extracellular factors and multiple feedback loops interact and regulate each other's' activity spatially and temporally to generate a gradient that patterns the embryo. The zebrafish system is sufficiently well defined now to allow effective and testable mathematical models to be generated that could move this field forward. Moreover, we know very little about how BMP regulators modulate the actual signaling gradient. The objective is to discover and discriminate mechanisms of BMP regulation by utilizing quantitative image acquisition and analysis, geometrically accurate mathematical models of early zebrafish embryo DV patterning, and mixed-quality constraint based optimization. In Aim 1 the spatiotemporal formation of the BMP signaling gradient will be quantitatively investigated by measuring phospho-Smad 1/5 levels IN TOTO in wild-type and BMP component mutant zebrafish embryos, segment all nuclei in each embryo, and register the data to a standard embryo. These studies will provide the first-ever (semi)-quantitative data that can be used to discern the spatial and quantitative differences and similarities of individual BMP extracellular modulators to understand their roles in BMP signaling gradient formation. In Aim 2 networks for BMP-mediated signaling control will be identified by developing, optimizing, and analyzing 3D spatiotemporal models. An image-based zebrafish late blastula- gastrula embryo BMP pattern formation model will be developed and tested for multiple alternative mechanisms of BMP regulation that guide pattern formation dynamics. In aim 3 we will use Model-Based Optimal Design of Experiments to reduce the complexity of factorial design required for comprehensive analysis of multiple-component networks. Additionally, we will determine the mechanism of Cvl2, Tsg1, and Chd regulation of dynamic BMP signaling to test the model's predictive ability and delineate the action of this important network. The goal of this aim is to carry out simultaneous gene perturbation experiments that will provide the greatest amount of information pertaining to BMP regulation. Understanding the mechanism of BMP-mediated patterning in vertebrates will provide the basis for tightly controlling BMP signaling in tissue regeneration and other prospective treatments of human disease.